Cancer statistics and the blame game

There's been a distinctly statistical flavour to this week, what with the question of lessons in gambling and The Moose's musings on child poverty.

I'll be the first to admit that, while I can more or less follow the reasoning of such Titans in the field as Mark Wadsworth (at least as long as he keeps explaining things so clearly), my grasp of the finer points is a little shakier than I'd like - I don't gamble, but if I did, I'd be the one standing in the bookies muttering "Each way? Er, is that r factorial divided by n minus r?"

Even I, however, have no difficulty working out from today's headlines that if just over 40% of cancers could be prevented by lifestyle change, then nearly 60% are determined by factors beyond the patient's control.

Unfortunately, this reasoning still seems to be beyond the grasp of some of those who work in the field. I make no apology, therefore, for recycling part of this post on the subject from last May, when similar statistics were published for breast cancer.

*******

The other 58% of cases may be linked to environmental or genetic factors or other causes not yet established. Information like this, however, proves a logical step too far for many NHS staff, for whom the mantra runs thus:

Cancer is caused by unhealthy lifestyles.
You have cancer.
ergo You have an unhealthy lifestyle.

In the past few years, several of my friends and family have been diagnosed with so-called ‘lifestyle cancers’, and, to a man (and woman) subjected to lengthy instruction by medical staff about their supposedly unhealthy habits despite a clear family history of the disease in each case.

Thus a friend who walks her dog several miles every day was advised to take more exercise; a lifelong non-drinker was repeatedly told to cut down on his alcohol consumption and, most bizarrely of all, a woman who has the healthiest diet I know of was constantly lectured on cutting down on fat and sugar and avoiding junk food – she weighs less than eight stone.

And each of these reported, with varying degrees of fury, a clear and consistent implication by hospital staff that they must have brought the cancer on themselves by their own failure to lead a healthy lifestyle. Their remonstrations were brushed aside - the cancer was proof enough.

It is no secret that doctors receive a ridiculously small amount of training in the interpretation of statistics, given the relevance of probabilities and incidence – I have mentioned before the GP who excused his diagnostic failure with the words, ‘97% of people with this cancer are obese; you aren’t even overweight, so there was only a 3% chance of you having it.’

That being so, how likely is it that the lower echelons of the medical hierarchy can correctly interpret statistical information, given the standard of maths in today's schools? It is a matter of record that numeracy skills are at a frighteningly low level across the population, and I doubt that hospital staff are any exception.

Tell them that cancer is linked to poor diet and lack of exercise and, unless it is clearly explained, some, at least, are going to go on with complete self-assurance to tell cancer patients that it is all their own fault.


Update: via Longrider, this BBC article includes an interesting show-trial interview with a woman brought out to make a public confession that her cancer was lifestyle-related.
Watch out for the interviewer posing the loaded question:
"Why was it you? What was it in your lifestyle that was wrong?"